I. INTRODUCTION. Calcific aortic stenosis (AS) is an increasingly common problem in developed countries, with an estimated prevalence of 2% to 4% among adults 65 years or older. Once symptoms develop, severe AS is associated with a median survival of approximately 2 years without intervention. Surgical aortic valve replacement (SAVR) has long been the standard of care for managing severe, symptomatic AS and can be done safely in the majority of patients. Historically, at least 30% of patients do not undergo SAVR owing to risk factors that are thought to put them at prohibitive surgical risk such as old age, prior cardiac surgeries, or other significant comorbidities. Unfortunately, the management of inoperable patients with medical therapy has done little to affect the natural history of the disease. Balloon aortic valvuloplasty (BAV) provides a short-term improvement in symptoms but a high rate of stenosis recurrence and no improvement in long-term survival without definitive valve replacement. In more recent years, transcatheter aortic valve replacement (TAVR) has provided a therapeutic option for many of these inoperable patients and is an alternative to SAVR in patients considered at moderate or high risk for complications with surgery.

A. TAVR is currently approved by the US Food and Drug Administration (FDA) and was granted the Conformité Européenne (CE) mark for high surgical risk and inoperable patients with severe AS deemed to be symptomatic from their valve disease with New York Heart Association (NYHA) functional class II or greater. For this indication, the US FDA has approved use of the balloon-expandable Edwards SAPIEN, SAPIEN-XT, and SAPIEN-3 (S3) valves (Edwards Life Sciences, Inc., Irvine, CA) and the self-expanding Medtronic CoreValve and CoreValve Evolut R (Medtronic, Inc., Minneapolis, MN). Patients considered at moderate risk for cardiac surgery have undergone TAVR with these valves as part of various clinical trials designed to test the safety and efficacy of this procedure among lower-risk groups. As a result of these trials, the S3 valve is approved for use in patients at intermediate surgical risk. The MCV is also approved for “valve-in-valve” TAVR for patients with severe bioprosthetic valve degeneration who are considered high risk or inoperable; the SAPIEN-XT is currently used for this application as part of a registry for inoperable patients. There are several additional devices that have obtained the CE mark of approval but are not currently FDA approved. These include the self-expandable Portico valve (St. Jude Medical, St. Paul, MN) and the uniquely designed Lotus Valve (Boston Scientific, Natick, MA) and Direct Flow Valve (Direct Flow Medical, Santa Rosa, CA), which is now discontinued and no longer available (Fig. 12.1).

B. The severity of calcific AS and appropriateness for intervention should be determined based on the recently published 2014 American Heart Association/American College of Cardiology Guidelines for the Management of Patients with Valvular Heart Disease. Surgical risk should be determined based on evaluation by a cardiac surgeon experienced in SAVR and should consider many factors including traditional tools such as the Society of Thoracic Surgeons predicted risk of mortality (STS-PROM) calculator or the European System for Cardiac Operative Risk Evaluation (Logistic EuroSCORE).
Certain factors that are not included in the formal risk calculators but are known to increase the risk for surgical mortality and/or morbidity include cirrhosis, prior radiation to the chest, prior bypass graft anatomy with an unsafe proximity of grafts to the chest wall to accommodate redo sternotomy, significant frailty, or a porcelain aorta.

A. TAVR is contraindicated in patients with a severely limited life expectancy owing to comorbid conditions, severe incapacitating dementia, or clear medical futility. The use of TAVR in patients with pure aortic regurgitation (AR) without stenosis is not currently approved although this indication is under study. Barriers to use in patients with isolated AR include the need to engineer devices that are designed to treat what is typically a dilated and noncalcified annulus. Noncalcified or minimally calcified valves may hold the device less securely and should be approached with caution. Some congenitally malformed valves are also not appropriate for TAVR.

B. TAVR is otherwise limited primarily by device availability and anatomic constraints. A very small or very large annulus may exceed the limits of the currently available devices. The most recently FDA-approved Edwards S3 devices are appropriate for annular sizes between 16 and 27 mm in diameter, and the CoreValve can be used for annular sizes between 18 and 30 mm according to manufacturer-published guidelines (Table 12.1).

C. The next anatomic limitation to TAVR relates to vascular access, and transfemoral (TF) TAVR may not be possible in some patients with small or severely calcified iliofemoral arteries. Generally, a minimal luminal diameter of 6 mm is required for most devices that utilize a #18 French sheath system, although the most recently approved Edwards S3 and Medtronic CoreValve Evolut R may be delivered through iliofemoral vessels as small as 5 mm in diameter depending on the degree of calcification present. Patients without a TF option can be considered for “alternative access,” which most often includes transapical (TA) or transaortic (TAo) approaches (Fig. 12.2). Other device delivery options include the carotid artery, subclavian artery, axillary artery, or femoral vein with either a trans-septal or transcaval (inferior vena cava to abdominal aorta) approach.

A. Evaluating the patient for high-risk aortic valve replacement (AVR) is a complex and multidisciplinary process termed the “Heart Team” approach that involves collaboration between experts in interventional cardiology, cardiac surgery, cardiovascular imaging, radiology, anesthesiology, and nursing to collectively determine the most appropriate strategy for each individual patient. It is also important to take into account the goals, expectations, and preferences of the patient. When considering TAVR, the first step is to determine surgical risk, and patients who are at low risk for SAVR should be offered a traditional valve replacement, or can be considered for enrollment in a randomized controlled trial such as the PARTNER III low-risk trial, if appropriate. When surgical risk is thought to be elevated or even prohibitive, further evaluation for transcatheter options is warranted.

B. The overwhelmingly preferred strategy for TAVR is via the TF approach, so the first step is to determine annular size and vascular access options. In patients with normal renal function, both factors can be quickly determined by a contrast-enhanced computed tomography (CT) scan. The CT of the chest should be gated (timed to the ECG so as to capture each image during the same portion of each cardiac cycle) to limit motion artifact and obtain the most accurate annular measurement. The annulus is frequently an elliptical structure, so it is important to determine size based on a three-dimensional imaging analysis that will take this asymmetry into account. Estimates of annular size based on two-dimensional echocardiography may underestimate the true annular size and lead to undersizing of the valve, a risk factor for paravalvular AR. The CT scan can then be continued through the abdomen and upper thigh to evaluate the iliofemoral system. When determining the appropriateness for TF-TAVR, it is important to consider not only the minimal luminal diameter but also the presence and extent of calcification and the tortuosity of the vessels.

C. When the patient’s renal function is not appropriate for a contrast-enhanced CT scan, alternative strategies for determining annular size include cardiac magnetic resonance imaging (MRI) and three-dimensional transesophageal echocardiography (TEE). In such cases, iliofemoral anatomy can be assessed by noncontrast CT scan. If further anatomic detail is required, CT of the iliofemoral system may be performed after positioning a pigtail catheter in the distal aorta under fluoroscopy and then transporting
the patient to the CT scanner for a direct, arterial contrast-injected CT scan using 10 to 15 mL of dye. Consideration may also be given to intravascular ultrasound evaluation of the pelvic vessels, though operators should be aware of the concern for a nonperpendicular image of the vessel and inaccurate sizing as a result.

D. In patients requiring alternative access, TA-TAVR is performed by puncturing the left ventricle (LV) apex through a small left anterior thoracotomy and crossing the aortic valve (AV) in an antegrade fashion. This technique has been limited to the balloon-expandable (SAPIEN) valves, which can be loaded onto the delivery device to accommodate either antegrade or retrograde implantation (Fig. 12.2). The unsheathing mechanism of the CoreValve, however, does not allow for the TA approach. TAo-TAVR is commonly performed via a partial sternotomy, although some operators prefer a right thoracotomy or manubriotomy. The delivery sheath is placed directly in the ascending aorta, followed by a traditional retrograde crossing of the AV, which can be performed with either balloon-expandable or self-expandable valves. Although both TA and TAo-TAVR require a more extensive surgical approach as compared to the TF approach, they do not require full sternotomy or cardiopulmonary bypass. In our experience, the TAo approach is less desirable in patients with a history of prior open heart surgery in whom mediastinal scarring can complicate the path to reexposing the ascending aorta. Thus, we prefer TA access for patients with prior coronary artery bypass grafting or valve surgery. The TAo approach is preferred, however, for most patients without prior surgery who do not have appropriate femoral access.

E. Alternative vascular access approaches have been extensively studied in hopes of finding safe and feasible options for TAVR in patients who lack appropriate TF access and who are not appropriate for TA or TAo-TAVR. Although these strategies are mostly limited to small series and case reports, operators have successfully completed TAVR from the carotid, subclavian, and axillary artery, as well as the femoral vein. Initial reports suggest that these approaches are both safe and feasible in selected patients. There are two possible approaches for completing TAVR from the femoral vein. The first involves passage of the catheter across the inter atrial septum, across the mitral valve, into the LV, and across the AV in an antegrade fashion. The second involves trans caval passage of the catheter from the inferior vena cava to the descending abdominal aorta, then over the aortic arch, and across the AV in a typical retrograde fashion. In transcaval cases, the arterio venous communication that is formed in the intra-abdominal retroperitoneal space is typically corrected with a closure device at the conclusion of the case.

F. In addition to determining access route, TAVR evaluation requires careful examination and management of other comorbid conditions, both cardiac and noncardiac. Patients with senile, calcific AS have a high incidence of atherosclerosis, and it is important to complete a left heart catheterization to evaluate for coronary artery disease. Pre-TAVR SYNTAX score (or residual SYNTAX score in patients who have undergone revascularization) has been shown to be a risk factor for cardiovascular mortality after TAVR in one prospective registry. Nevertheless, there are no randomized trials guiding the decision for revascularization. Although the initial trials of TAVR required full revascularization, in current clinical practice, usually only patients with a large area of ischemia, typical angina symptoms, or high-grade stenosis in an “important” anatomic location are likely to undergo percutaneous coronary intervention (PCI) prior to TAVR. Anecdotally, conservative medical management of certain asymptomatic but angiographically severe lesions has not compromised the success of TAVR in selected patients at our institution.

G. Carotid artery disease is also important to consider prior to TAVR. Patients with symptomatic, high-grade lesions should likely undergo carotid endarterectomy or stenting prior to TAVR. Intermediate and asymptomatic lesions do not necessitate prophylactic intervention, but special consideration should be given to unprotected carotid artery stenosis (significant stenosis with complete occlusion of the contralateral internal carotid artery), or patients who may be at an especially high risk for prolonged hypotension during TAVR.

H. Conduction system disease must be considered as well, especially as it pertains to device selection. One drawback of the self-expandable valves has been a significantly higher incidence of heart block requiring permanent pacemaker (PPM) implantation as compared to the balloon-expandable valves (˜28% vs. 6%, respectively) and may be a deciding factor in valve selection in certain patients. As with SAVR, the left bundle of His is especially vulnerable to disruption after TAVR, so patients with preexisting right bundle branch block (RBBB) or bifascicular block are at especially high risk for post-TAVR complete heart block. If PPM implantation is required, placement of a biventricular pacemaker may be considered in patients with a reduced ejection fraction.